← Back to Home

Overview

FRC Splat is a data-driven simulator and dashboard for forecasting district points, DCMP/Worlds qualification, and event outcomes across the FRC season. Note that all predictions on this site (aside from offseasons) are for NEXT year. Until new event rosters are filled out, we will run simulations assuming the same rosters (and event codes) as 2025. Below is a quick tour of what you can do here, plus details on how the simulator works under the hood.

What you can do on this site

Search everything from the home page. Results are ordered Districts → District Events → Regionals → Teams. Start typing a district (e.g., chs), an event code (e.g., 2025vagle), or a team number.
District Point Forecasts. See team-level forecasts for district points, DCMP qualification percent, and Worlds odds by district. (districts.html)
Projected District Point Cutoffs. View projected cutoffs for DCMP and Worlds, updated as simulations refresh. (cutoffs.html)
District Event Predictions. For each event, inspect per-team placement probabilities and district point distributions. When an event finishes, the page detects completion and shows a banner This event has been completed. (events.html)
Regional Event Predictions. Similar forecasts for regional events. (regionals.html)
Team pages with event shortcuts. Each team page shows the DCMP and Worlds Qualification Forecasts for the team as well as cards linking directly to every event that team attends. (team.html)
Methodology & Changelog. Read the modeling details and follow updates. (methodology.html, changelog.html)

Go to

District Point Forecasts

Go to

Projected Cutoffs

Go to

District Event Predictions

Go to

Regional Event Predictions

Looking Into the Machine: How SPLAT (mostly) works

District points by team. Distribution of district points for each team across 10,000 simulations.
- Empirically learn the distribution of end-of-season EPA given prior-season EPA.
- Use Statbotics formulae for next season’s starting EPA.
- Path from start to end EPA uses random “steps,” with normal noise controlling deviation from straight-line progression.
At each event, qualification match schedules are generated to reflect FIRST scheduler priorities.
- Minimize repeat teammates/opponents; do not attempt to equalize time between matches.
- Repeat teammates/opponents incur an exponentially increasing penalty.
At each event, teams play to their simulated end-of-event EPA over their qual matches.
- Each simulation has a unique schedule and random variance in team strength.
- Uses overall EPA and RP-component strength metrics.
Alliance selections are simulated after quals.
- Each team gets a desirability score (EPA + small random variance); captains pick the top remaining team.
- Declines modeled with projecting forward to potential partnerships and comparing EPA sums.
Elims use the standard double-elimination format with best-of-3 finals.
- Alliance EPAs drive match predictions; assume no backups used.
Awards are based on rank and award history (last 2 years).
- 5-point awards weighted by rank/history. EI/Impact put special emphasis on prior culture award winners over the last two years (weights learned by neural network). RAS is randomly distributed among eligible rookies.
District points are tabulated based on each team’s first two in-district events; Impact winners plus top remaining by points advance to DCMP.
DCMP fields are assigned at random (if multiple fields); each field counts as a separate event for points purposes.
Worlds invitations include qualifying award winners, event winners (with specific exceptions), HOF teams, Worlds Impact Finalists, Worlds Winners, and next-highest district point scorers.